Here is some additional information to help you prepare for the AI-900 or can be used just to solidify your knowledge of these concepts.

Machine Learning Techniques Comparison Table

Aspect	Regression	Classification	Clustering
Type of Learning	Supervised	Supervised	Unsupervised
Primary Goal	Predict a numeric value	Predict a category or label	Group similar data points
Output Type	Continuous number	Discrete category	Cluster/group assignment
Labeled Training Data	Yes	Yes	No
Key Question Answered	How much? How many? How long?	Which category? Yes or No?	Which items are similar?
Common Keywords	Predict, estimate, forecast	Classify, assign, detect	Group, segment, organize
Typical Output Examples	Price, temperature, revenue, time	Approved/Rejected, Spam/Not spam	Customer segments, usage groups
Example Scenario	Predict house prices	Detect fraudulent transactions	Segment customers by behavior
AI-900 Exam Focus	Identifying numeric predictions	Identifying label predictions	Identifying pattern discovery
Common Exam Trap	Confusing ranges with categories	Treating Yes/No as numeric	Assuming labels exist

---

Quick Visual Memory Trick

• Regression → 📈 Numbers on a line
• Classification → 🏷️ Named buckets
• Clustering → 🧩 Natural groupings

---

Side-by-Side Example

Imagine a retail company:

Business Question	Technique
“What will next month’s revenue be?”	Regression
“Will this customer churn?”	Classification
“Which customers behave similarly?”	Clustering

---

Common AI-900 Exam Pitfalls to Avoid

• ❌ High / Medium / Low → Classification, not regression
• ❌ Yes / No → Classification, not regression
• ❌ Grouping without predefined labels → Clustering
• ❌ Predicting quantities → Regression

---

Exam-Day Decision Shortcut

Ask yourself one question:

“Is the output a number?”

• Yes → Regression
• No, it’s a label → Classification
• No labels, just groups → Clustering

---

Go to the AI-900 Exam Prep Hub main page.